Examples with TrpTextRegionType eu.transkribus.core.model.beans.pagecontent_trp.TrpTextRegionType used on opensource projects

Example 11 with TrpTextRegionType

use of eu.transkribus.core.model.beans.pagecontent_trp.TrpTextRegionType in project TranskribusCore by Transkribus.

the class TrpXlsxBuilder method writeXlsxForDoc.

public static void writeXlsxForDoc(TrpDoc doc, boolean wordBased, File exportFile, Set<Integer> pageIndices, IProgressMonitor monitor, ExportCache cache) throws NoTagsException, Exception {
    if (cache == null) {
        throw new IllegalArgumentException("ExportCache must not be null.");
    }
    if (cache.getCustomTagMapForDoc().isEmpty()) {
        logger.info("No tags to store -> Xlsx export cancelled");
        throw new NoTagsException("No tags available to store into Xlsx");
    }
    List<TrpPage> pages = doc.getPages();
    String exportPath = exportFile.getPath();
    Set<String> selectedTags = cache.getOnlySelectedTagnames(ExportUtils.getOnlyWantedTagnames(CustomTagFactory.getRegisteredTagNames()));
    int totalPages = pageIndices == null ? pages.size() : pageIndices.size();
    if (monitor != null) {
        monitor.beginTask("Exporting to Excel", totalPages);
    }
    wb = new XSSFWorkbook();
    int c = 0;
    for (int i = 0; i < pages.size(); ++i) {
        if (pageIndices != null && !pageIndices.contains(i))
            continue;
        if (monitor != null) {
            if (monitor.isCanceled()) {
                throw new InterruptedException("Export was canceled by user");
            // logger.debug("Xlsx export cancelled!");
            // return;
            }
            monitor.subTask("Processing page " + (c + 1));
        }
        TrpPage page = pages.get(i);
        // try to get previously loaded JAXB transcript
        JAXBPageTranscript tr = null;
        if (cache != null) {
            tr = cache.getPageTranscriptAtIndex(i);
        }
        if (tr == null) {
            TrpTranscriptMetadata md = page.getCurrentTranscript();
            tr = new JAXBPageTranscript(md);
            tr.build();
        }
        // old version
        // TrpPage page = pages.get(i);
        // TrpTranscriptMetadata md = page.getCurrentTranscript();
        // JAXBPageTranscript tr = new JAXBPageTranscript(md);
        // tr.build();
        TrpPageType trpPage = tr.getPage();
        logger.debug("writing xlsx for page " + (i + 1) + "/" + doc.getNPages());
        List<TrpTextRegionType> textRegions = trpPage.getTextRegions(true);
        for (int j = 0; j < textRegions.size(); ++j) {
            TrpTextRegionType r = textRegions.get(j);
            List<TextLineType> lines = r.getTextLine();
            for (int k = 0; k < lines.size(); ++k) {
                TrpTextLineType trpL = (TrpTextLineType) lines.get(k);
                List<WordType> words = trpL.getWord();
                if (wordBased) {
                    for (int l = 0; l < words.size(); ++l) {
                        TrpWordType w = (TrpWordType) words.get(l);
                        writeTagsForShapeElement(w, trpL.getUnicodeText(), String.valueOf(doc.getId()), String.valueOf(page.getPageNr()), r.getId(), trpL.getId(), w.getId(), selectedTags);
                    }
                } else {
                    writeTagsForShapeElement(trpL, trpL.getUnicodeText(), String.valueOf(doc.getId()), String.valueOf(page.getPageNr()), r.getId(), trpL.getId(), "", selectedTags);
                }
            }
        }
        ++c;
        if (monitor != null) {
            monitor.worked(c);
        }
    }
    /*
		 * auto size the columns
		 */
    for (int i = 0; i < wb.getNumberOfSheets(); i++) {
        int numberOfCells = 0;
        Iterator rowIterator = wb.getSheetAt(i).rowIterator();
        /**
         * Escape the header row *
         */
        if (rowIterator.hasNext()) {
            Row headerRow = (Row) rowIterator.next();
            // get the number of cells in the header row
            numberOfCells = headerRow.getPhysicalNumberOfCells();
            for (int j = 0; j < numberOfCells; j++) {
                wb.getSheetAt(i).autoSizeColumn(j);
            }
        }
    }
    FileOutputStream fOut;
    try {
        // means no tags at all
        if (wb.getNumberOfSheets() == 0) {
            throw new IOException("Sorry - No tags available for export");
        }
        fOut = new FileOutputStream(exportPath);
        wb.write(fOut);
        fOut.close();
    } catch (IOException e) {
        // TODO Auto-generated catch block
        e.printStackTrace();
        throw e;
    }
    logger.info("wrote xlsx to: " + exportPath);
}

Example 12 with TrpTextRegionType

use of eu.transkribus.core.model.beans.pagecontent_trp.TrpTextRegionType in project TranskribusCore by Transkribus.

the class TrpTeiStringBuilder method setContent.

@Override
protected void setContent(List<TrpPage> pages) throws JAXBException, InterruptedException {
    SebisStringBuilder sbFacsimile = new SebisStringBuilder();
    SebisStringBuilder sbText = new SebisStringBuilder();
    sbText.incIndent();
    sbText.addLine("<text>");
    sbText.incIndent();
    sbText.addLine("<body>");
    // sbText.incIndent();
    // text = tei.createElementNS(TEI_NS, "text");
    // body = tei.createElementNS(TEI_NS, "body");
    int totalPages = pageIndices == null ? pages.size() : pageIndices.size();
    if (monitor != null) {
        monitor.beginTask("Creating TEI", totalPages);
    }
    int c = 0;
    for (int i = 0; i < pages.size(); ++i) {
        if (pageIndices != null && !pageIndices.contains(i))
            continue;
        if (monitor != null) {
            if (monitor.isCanceled()) {
                throw new InterruptedException("Export was canceled by user");
            // break;
            }
            monitor.subTask("Processing page " + (c + 1));
        }
        TrpPage p = pages.get(i);
        logger.debug("1Processing page " + p.getPageNr() + ": " + p.getUrl() + " - XML=" + p.getCurrentTranscript().getUrl());
        // check buffer for transcript or unmarshal the page XML
        PcGtsType pc = this.getPcGtsTypeForPage(p);
        if (pars.hasZones()) {
            // create a facsimile element for each page that are appended to the root element of the TEI after header
            openFacsimileElement(sbFacsimile, p, pc);
        }
        // create page-break element for each page as child of body element:
        writePageBreak(sbText, p, pc);
        // 
        // // append all text-regions / lines / words to the xml:
        List<TrpRegionType> regions = pc.getPage().getTextRegionOrImageRegionOrLineDrawingRegion();
        Collections.sort(regions, new TrpElementReadingOrderComparator<RegionType>(true));
        for (TrpRegionType r : regions) {
            // System.out.println(r.getClass());
            if (r instanceof TextRegionType) {
                if (pars.hasZones()) {
                    writeZonesForTextRegion(sbFacsimile, (TrpTextRegionType) r, p.getPageNr());
                }
                writeTextForTextRegion(sbText, (TrpTextRegionType) r, p.getPageNr());
            } else {
                // write other regions
                if (pars.hasZones()) {
                    String facsId = FACS_ID_PREFIX + p.getPageNr();
                    writeZoneForShape(sbFacsimile, r, facsId, true);
                }
            }
        }
        if (pars.hasZones()) {
            closeFacsimilieElement(sbFacsimile);
        }
        ++c;
        if (monitor != null) {
            monitor.worked(c);
        }
    }
    // text.appendChild(body);
    // root.appendChild(text);
    // sbText.decIndent();
    sbText.addLine("</body>");
    sbText.decIndent();
    sbText.addLine("</text>");
    sbText.decIndent();
    sbTotal.sb.append(sbFacsimile.toString());
    sbTotal.sb.append(sbText.toString());
}

Example 13 with TrpTextRegionType

use of eu.transkribus.core.model.beans.pagecontent_trp.TrpTextRegionType in project TranskribusCore by Transkribus.

the class DocxBuilder method writeDocxForTranscriptWithTables.

private static void writeDocxForTranscriptWithTables(MainDocumentPart mdp, TrpPageType trpPage, boolean wordBased, boolean preserveLineBreaks) {
    boolean rtl = false;
    // TrpTableRegionType is contained in the regions too
    List<TrpRegionType> regions = trpPage.getRegions();
    Collections.sort(regions, new TrpElementReadingOrderComparator<RegionType>(true));
    for (int j = 0; j < regions.size(); ++j) {
        TrpRegionType r = regions.get(j);
        if (r instanceof TrpTableRegionType) {
            logger.debug("is table");
            TrpTableRegionType table = (TrpTableRegionType) r;
            int cols = table.getNCols();
            int rows = table.getNRows();
            // PageXmlUtils.buildPolygon(table.getCoords().getPoints()).getBounds().getMaxX();
            double maxX = table.getBoundingBox().getMaxX();
            // PageXmlUtils.buildPolygon(table.getCoords().getPoints()).getBounds().getMinX();
            double minX = table.getBoundingBox().getMinX();
            int tablesize = (int) (maxX - minX);
            List<List<TrpTableCellType>> allRowCells = new ArrayList<List<TrpTableCellType>>();
            for (int k = 0; k < rows; k++) {
                allRowCells.add(table.getRowCells(k));
            }
            List<HashMap<Integer, TrpTableCellType>> allRows = new ArrayList<HashMap<Integer, TrpTableCellType>>();
            HashMap<Integer, TrpTableCellType> nextRowMap = new HashMap<Integer, TrpTableCellType>();
            for (List<TrpTableCellType> rowCells : allRowCells) {
                HashMap<Integer, TrpTableCellType> currRowMap = new HashMap<Integer, TrpTableCellType>();
                /*
	            	 * fill up all cells which are not set in TRP (needed for vertical cell merge)
	            	 * the nextRowMap contains already all cells which span vertically with the cells above - means they got merged 
	            	 * in the table but have to be considered here 
	            	 */
                currRowMap.putAll(nextRowMap);
                nextRowMap.clear();
                for (TrpTableCellType cell : rowCells) {
                    // logger.debug("table cell text " + cell.getUnicodeTextFromLines());
                    currRowMap.put(cell.getCol(), cell);
                    if (cell.getRowSpan() > 1) {
                        nextRowMap.put(cell.getCol(), null);
                    }
                }
                allRows.add(currRowMap);
            }
            Tbl thisTable;
            try {
                thisTable = getDocxTable(wordMLPackage, wordBased, rows, cols, allRows, tablesize, mdp);
                mdp.addObject(thisTable);
            } catch (Exception e) {
                // TODO Auto-generated catch block
                e.printStackTrace();
            }
            // this Br element is used break the current and go for next line
            Br br = factory.createBr();
            org.docx4j.wml.P p = factory.createP();
            mdp.addObject(p);
            p.getContent().add(br);
        } else if (r instanceof TrpTextRegionType) {
            TrpTextRegionType tr = (TrpTextRegionType) r;
            /*
				 * create one paragraph for each text region
				 * but only if there is some text in it
				 */
            String helper = tr.getUnicodeText().replaceAll("\n", "");
            if (!helper.equals("")) {
                exportTextRegion(tr, wordBased, null, mdp);
            }
        }
    }
}

Example 14 with TrpTextRegionType

use of eu.transkribus.core.model.beans.pagecontent_trp.TrpTextRegionType in project TranskribusCore by Transkribus.

the class TrpRtfBuilder method getRtfParagraphsForTranscript.

// public static void writeRtfForElement(Rtf rtf, ITrpShapeType element, boolean wordBased, File file, boolean append) throws IOException, JAXBException {
// element.getUnicodeText();
// CustomTagList cl = element.getCustomTagList();
// 
// RtfText text = RtfText.text(element.getUnicodeText());
// text = formatRtfText(text, element.getTextStyle());
// 
// 
// 
// if (element instanceof TextLineType || element instanceof TextRegionType) {// TODO words vs lines and regions
// rtf.p(text);
// } else if (element instanceof TrpWordType) {
// //			rtf.p(texts);
// }
// 
// 
// //		cl.getCustomTagAndContinuations(tag)
// 
// 
// 
// }
public static RtfPara[] getRtfParagraphsForTranscript(TrpPageType trpPage, boolean wordBased) throws IOException, JAXBException {
    boolean rtl = false;
    List<TrpTextRegionType> textRegions = trpPage.getTextRegions(true);
    // List<TrpTextRegionType> textRegions = trpPage.getTextRegionsAndTextRegionsFromTableRegions(true);
    RtfPara[] paras = new RtfPara[textRegions.size()];
    for (int j = 0; j < textRegions.size(); ++j) {
        TrpTextRegionType r = textRegions.get(j);
        // if (exportTags){
        // getTagsForShapeElement(r);
        // }
        List<TextLineType> lines = r.getTextLine();
        RtfText[] linesTexts = new RtfText[lines.size()];
        for (int i = 0; i < lines.size(); ++i) {
            TrpTextLineType trpL = (TrpTextLineType) lines.get(i);
            linesTexts[i] = (wordBased && trpL.getWord().size() > 0) ? getRtfTextForLineFromWords(trpL) : getRtfTextForShapeElement(trpL);
            linesTexts[i] = RtfText.text(linesTexts[i], "\n");
        }
        // read from right to left -> alignment is right
        if (rtl) {
        // paras[j] = RtfPara.p(linesTexts).footnote("Test").alignRight();
        } else {
            String test = "test";
            paras[j] = RtfPara.p(linesTexts);
        // paras[j] = RtfPara.p(linesTexts, RtfText.footnote("Test")).alignLeft();
        }
    }
    return paras;
// Rtf rtf = Rtf.rtf().section(paras);
// return rtf;
// for (RegionType r : trpPage.getTextRegionOrImageRegionOrLineDrawingRegion()) {
// if (r instanceof GraphicRegionType) {
// GraphicRegionType gr = (GraphicRegionType) r;
// // TODO: how to export images in pdf??
// r.getTextRegions(recursive);
// }
// }
// tr.getPage().getTextRegions(recursive);
// Rtf.rtf();
// RtfWriter;
}

Example 15 with TrpTextRegionType

use of eu.transkribus.core.model.beans.pagecontent_trp.TrpTextRegionType in project TranskribusCore by Transkribus.

the class TrpTxtBuilder method writeTxtForSinglePage.

private static void writeTxtForSinglePage(File file, TrpPageType trpPage, boolean wordBased, boolean preserveLineBreaks) {
    boolean rtl = false;
    // TrpTableRegionType is contained in the regions too
    List<TrpRegionType> regions = trpPage.getRegions();
    Collections.sort(regions, new TrpElementReadingOrderComparator<RegionType>(true));
    List<String> content = new ArrayList<String>();
    for (int j = 0; j < regions.size(); ++j) {
        TrpRegionType r = regions.get(j);
        if (r instanceof TrpTableRegionType) {
            /*
				 * TODO: for simple txt export: how to handle tables
				 */
            continue;
        } else if (r instanceof TrpTextRegionType) {
            TrpTextRegionType tr = (TrpTextRegionType) r;
            List<TextLineType> lines = tr.getTextLine();
            for (int i = 0; i < lines.size(); ++i) {
                TrpTextLineType trpL = (TrpTextLineType) lines.get(i);
                String textOfCurrLine = trpL.getUnicodeText();
                if (wordBased && trpL.getWord().size() > 0) {
                    for (WordType word : trpL.getWord()) {
                        content.add(((ITrpShapeType) word).getUnicodeText());
                    }
                } else if (textOfCurrLine != "") {
                    content.add(textOfCurrLine);
                }
            // if(preserveLineBreaks){
            // content.add(System.lineSeparator());
            // }
            }
            if (lines.size() > 0) {
                content.add(System.lineSeparator());
            // try {
            // //Add line separator after each region
            // Files.write(Paths.get(file.getAbsolutePath()), new ArrayList<String>() {{ add(System.lineSeparator()); }}, utf8,
            // StandardOpenOption.CREATE, StandardOpenOption.APPEND);
            // } catch (IOException e) {
            // // TODO Auto-generated catch block
            // e.printStackTrace();
            // }
            }
        }
    }
    try {
        logger.debug("path " + Paths.get(file.getAbsolutePath()));
        Files.write(Paths.get(file.getAbsolutePath()), content, utf8, StandardOpenOption.CREATE, StandardOpenOption.APPEND);
    } catch (IOException e) {
        // TODO Auto-generated catch block
        e.printStackTrace();
    }
}